基于深度学习的图像局部模糊识别

doi:10.3969/j.issn.0255-8297.2018.02.011

应用科学学报 ›› 2018, Vol. 36 ›› Issue (2): 321-330.doi: 10.3969/j.issn.0255-8297.2018.02.011

基于深度学习的图像局部模糊识别

杨滨^1,4, 张涛², 陈先意³

1. 江南大学设计学院, 江苏无锡 214122;
2. 江南大学物联网学院, 江苏无锡 214122;
3. 南京信息工程大学计算机与软件学院, 南京 210044;
4. 江南大学轻工过程先进控制教育部重点实验室, 江苏无锡 214122

收稿日期:2018-02-06 出版日期:2018-03-31 发布日期:2018-03-31
作者简介:杨滨,副教授,研究方向:图像处理,机器学习,E-mail:Yangbin@jiangnan.edu.cn
基金资助:
国家自然科学基金（No.61232016，No.61502242）资助

Local Blur Detection of Digital Images Based on Deep Learning

YANG Bin^1,4, ZHANG Tao², CHEN Xian-yi³

1. School of Design, Jiangnan University, Wuxi 214122, Jiangsu Province, China;
2. School of Internet of Things, Jiangnan University, Wuxi 214122, Jiangsu Province, China;
3. School of Computer and Software, Nanjing University of Information Science and Technology, Nanjing 210044, China;
4. Key Laboratory of Advanced Process Control of Light Industry of Ministry of Education, Jiangnan University, Wuxi 214122, Jiangsu Province, China

Received:2018-02-06 Online:2018-03-31 Published:2018-03-31

摘要/Abstract

摘要：

数字图像中常用模糊操作隐藏或抹去篡改的痕迹.为此，针对常用的高斯模糊、均值模糊及中值模糊操作的识别问题，构建了一种卷积神经网络模型，并给出其网络拓扑结构.在传统的卷积神经网络模型中添加一个信息处理层，提取出输入图像块的滤波频域残差特征，以提高网络模型对一次滤波与二次滤波操作的识别性.实验结果表明，所提方法的准确率较以往传统方法有较大提升，且泛化性能优越，能检测出主流的线性和非线性滤波操作。

关键词: 滤波检测, 模糊识别, 深度学习, 卷积神经网络, 图像取证

Abstract:

Blurring is generally a post-operation to conceal or remove the trace of tampering. In this paper, a new convolutional neural network model is proposed, and the corresponding network topology is presented to handle the problems in the detection of blur operations, such as Gaussian blur, average blur, or median blur. An information process layer is added into the conventional convolutional neural network to extract the residual features of fltering frequency domain, accordingly, improving the accuracy of blur detection between the frst-order and the second-order fltering operations. Experimental results demonstrate that the proposed method performs a higher accuracy in blur detection than traditional methods, and is able to discriminate between the common linear and nonlinear blur operations.

Key words: deep learning, image forensic, blur detection, convolutional neural network, fltering detection

中图分类号:

TP309

杨滨, 张涛, 陈先意. 基于深度学习的图像局部模糊识别[J]. 应用科学学报, 2018, 36(2): 321-330.

YANG Bin, ZHANG Tao, CHEN Xian-yi. Local Blur Detection of Digital Images Based on Deep Learning[J]. Journal of Applied Sciences, 2018, 36(2): 321-330.

参考文献

[1] Yang B, Sun X M, Guo H L, Xia Z H, Chen X Y. A copy-move forgery detection method based on CMFD-SIFT[J]. Multimedia Tools and Applications, 2018, 77(1):837-855.
[2] Kirchner M, Fridrich J. On detection of median fltering in digital images[J]. Media Forensics & Security Ⅱ, 2010, 7541(1):754110.
[3] 彭安杰,康显桂. 基于滤波残差多方向差分的中值滤波取证技术[J]. 计算机学报,2016, 39(3):503-515. Peng A J, Kang X G. Median fltering forensics based on multi-directional difference of fltering residuals[J]. Chinese Journal of Computers, 2016, 39(3):503-515. (in Chinese)
[4] 李明富,王晨,彭安杰,陈立伟. 数字图像模糊滤波盲取证算法[J]. 半导体光电,2017, 38(3):430-434. Li M F, Wang C, Peng A J, Chen L W. A blind forensics algorithm for digital image smoothing fltering[J]. Semiconductor Optoelectronics, 2017, 38(3):430-434. (in Chinese)
[5] Dixit R, Naskar R. Review, analysis and parameterization of techniques for copy-move forgery detection in digital images[J]. IET Image Processing, 2017, 11(9):746-759.
[6] Chen J, Kang X, Liu Y, Wang Z J. Median fltering forensics based on convolutional neural networks[J]. IEEE Signal Processing Letters, 2015, 22(11):1849-1853.
[7] Gu B, Sun X, Sheng V S. Structural minimax probability machine[J]. IEEE Transactions on Neural Networks & Learning Systems, 2017, 28(7):1646-1656.
[8] Bayar B, Stamm M C. A deep learning approach to universal image manipulation detection using a new convolutional layer[J]. ACM Workshop on Information Hiding and Multimedia Security, 2016:5-10.
[9] Liu A, Zhao Z, Zhang C, Su Y. Smooth fltering identifcation based on convolutional neural networks[J]. Multimedia Tools & Applications, 2016:1-15.
[10] 周飞燕,金林鹏,董军. 卷积神经网络研究综述[J]. 计算机学报,2017, 40(6):1229-1251. Zhou F Y, Jin L P, Dong J. Review of convolutional neural network[J]. Chinese Journal of Computers, 2017, 40(6):1229-1251. (in Chinese)
[11] Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R. Dropout:a simple way to prevent neural networks from overftting[J]. Journal of Machine Learning Research, 2014, 15(1):1929-1958.
[12] Bas P, Filler T, Pevný T. "Break our steganographic system":the ins and outs of organizing BOSS[J]. Journal of the American Statistical Association, 2011, 96(454):488-499.
[13] Schaefer G. UCID:an uncompressed color image database[C]//Storage & Retrieval Methods & Applications for Multimedia. DBLP, 2003:472-480.
[14] Dong J, Wang W, Tan T. CASIA image tampering detection evaluation database[C]//IEEE China Summit & International Conference on Signal and Information Processing, 2013:422-426.
[15] Xu J, Ling Y, Zheng X. Forensic detection of Gaussian low-pass fltering in digital images[C]//International Congress on Image and Signal Processing, 2015:819-823.

基于深度学习的图像局部模糊识别

Local Blur Detection of Digital Images Based on Deep Learning

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	王孟轩, 张胜, 王月, 雷霆, 杜渂. 改进的CRNN模型在警情文本分类中的研究与应用[J]. 应用科学学报, 2020, 38(3): 388-400.
[2]	马鑫, 吴云, 鹿泽光. 基于混合神经网络的协同过滤推荐模型[J]. 应用科学学报, 2020, 38(3): 478-487.
[3]	刘伟, 章琬苓, 项世军. 基于LBP-MDCT和CNN的人脸活体检测算法[J]. 应用科学学报, 2019, 37(5): 609-617.
[4]	王灿军, 廖鑫, 陈嘉欣, 秦拯, 刘绪崇. 基于卷积神经网络的面部图像修饰检测[J]. 应用科学学报, 2019, 37(5): 618-630.
[5]	吴韵清, 吴鹏, 陈北京, 鞠兴旺, 高野. 基于残差全卷积网络的图像拼接定位算法[J]. 应用科学学报, 2019, 37(5): 651-662.
[6]	朱翌明, 陈帆, 和红杰, 陈鸿佑. 基于秘密信息驱动的正交GAN信息隐藏模型[J]. 应用科学学报, 2019, 37(5): 721-732.
[7]	靳华中, 刘潇龙, 胡梓珂. 一种结合全局和局部特征的图像描述生成模型[J]. 应用科学学报, 2019, 37(4): 501-509.
[8]	赵云山, 段友祥. 基于Attention机制的卷积神经网络文本分类模型[J]. 应用科学学报, 2019, 37(4): 541-550.
[9]	李芳, 钟映春, 戚剑, 罗鹏. 周围神经MicroCT图像中神经束区域的自动分离[J]. 应用科学学报, 2019, 37(3): 359-368.
[10]	刘佳睿, 卢伟, 刘轲, 黄信朝, 蔺聪, 刘先进. 基于超像素分割的图像复制粘贴篡改检测[J]. 应用科学学报, 2019, 37(3): 419-426.
[11]	曾润华, 张树群. 改进卷积神经网络的语音情感识别方法[J]. 应用科学学报, 2018, 36(5): 837-844.
[12]	史晓裕, 李斌, 谭舜泉. 深度学习空域隐写分析的预处理层[J]. 应用科学学报, 2018, 36(2): 309-320.
[13]	董伟, 王建军. 改进的卷积神经网络用于对比度增强取证[J]. 应用科学学报, 2017, 35(6): 745-753.
[14]	张晓琳，方针，张新鹏. 利用通道间相关性的CFA图像盲取证[J]. 应用科学学报, 2015, 33(1): 87-94.
[15]	李晟;张新鹏. 利用JPEG压缩特性的合成图像检测[J]. 应用科学学报, 2008, 26(3): 281-281 .